latishab/turnsense
A lightweight end-of-utterance detection model fine-tuned on SmolLM2-135M, optimized for Raspberry Pi and low-power devices.
Turnsense helps voice AI developers create more natural conversational experiences by accurately detecting when a user has finished speaking. It takes the text output from a speech-to-text system and determines if a full thought has been expressed, allowing the AI to respond at the right moment. This is ideal for developers building voice assistants or interactive voice applications on resource-constrained devices like Raspberry Pi.
Use this if you are building real-time voice AI applications for low-power edge devices and need a reliable way to know when a user's spoken turn has ended.
Not ideal if your speech-to-text system frequently produces text without proper punctuation or struggles with very short, ambiguous utterances.
Stars
46
Forks
2
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Nov 08, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/latishab/turnsense"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
IBM/MAX-Text-Sentiment-Classifier
Detect the sentiment captured in short pieces of text
giuseppebonaccorso/twitter_sentiment_analysis_word2vec_convnet
Twitter Sentiment Analysis with Gensim Word2Vec and Keras Convolutional Network
pabitralenka/Customer-Feedback-Analysis
Multi Class Text (Feedback) Classification using CNN, GRU Network and pre trained Word2Vec...
danielegrattarola/twitter-sentiment-cnn
An implementation in TensorFlow of a convolutional neural network (CNN) to perform sentiment...
hpanwar08/sentiment-analysis-torchtext
Seniment Analysis in Torchtext